A Dose Reconstruction Analysis for Determining Optimal Treatment Margins for Image-Guided and Daily Adaptive Radiation Therapy

Abstract

Purpose/Objective: The utilization of daily Image-Guided Radiation Therapy (IGRT) for the reduction of treatment margins in prostate cancer has been the subject of intense research efforts. Bowel gas, feces, and urine can cause deformation of the prostate and/or seminal vesicles that requires an unknown amount of margin. The purpose of this retrospective study was 1.) Perform dose reconstruction for thirty prostate patients imaged and treated daily with helical tomotherapy and 2.) Calculate the minimum treatment margins required for IGRT treatment delivery. Materials/Methods: A series of retrospective treatment plans were created for 30 prostate patients. The gross tumor volume (GTV) for the first 50 Gy was defined as the entire prostate gland plus the seminal vesicles. A series of planning target volumes (PTVs) were created with gradually decreasing margins: 10-mm/5-mm (5-mm posterior, and 10-mm in all other directions), 8-mm/4-mm, 6-mm/3-mm, 4-mm/2-mm, and 2-mm/0-mm. Archived Megavoltage CT (MVCT) images were used in this dose reconstruction analysis. The MVCT images were initially acquired prior to treatment delivery for CT-guided localization and patient repositioning. Retrospectively, the prostate, seminal vesicles, and rectum were manually contoured on each MVCT image. The delivered doses were then recalculated using the treatment delivery sequence from each of the five treatment plans. The smallest margins required for 100 percent of the prostate or seminal vesicles to be covered by the prescribed dose for 95 percent of the delivered fractions were measured for each patient. Results: The smallest prostate margins for all patients was 2-mm posterior and 4-mm in all other directions (4-mm/2-mm). The prostate was slightly deformed in patients with excessive bowel gas. However, the CT-guided localization and patient repositioning software on the delivery system generally compensated for this deformation. In contrast, the seminal vesicles were substantially more deformable. The smallest seminal vesicle margins were 6-mm/3-mm for 66% of the patients, 8-mm/4-mm for 31% of the patients, and greater than 10-mm/5-mm for one patient. This patient had a rectal diameter that varied from 2 to 8-cm during the course of treatment. During fractions where the rectum was extremely full, the seminal vesicles were severely deformed relative to the initial treatment planning CT. Ideally, a library of treatment plans with a similar range of margins could be utilized prospective at the time of treatment. This simple form of daily adaptive therapy could be easily implemented by prospectively selecting the smallest margins for each fraction. The cost of performing this type of adaptive therapy is the time requirement for creating the additional treatments plans. Tools for automating the creation of the reduced margin plans are currently being developed. Conclusions: A dose reconstruction analysis has been performed for patients treated with CT based IGRT. Treatment margins of 6-mm/3-mm would have been acceptable for 66% of the analyzed patients. Patients with large fluctuations in bowel gas will require larger planning margins, even with IGRT.